Message242575
As for "why not length instead of zeroes": Because the primary reason for the parameter is specifying that the string can contain embedded zeroes. Returning the length is a side-effect of this, not the main point. If the string didn't have embedded zeroes, we wouldn't need the length.
The only reason the code didn't have "zeroes=True" everywhere was because I screwed up and didn't realize all those mappings *should* have specified it.
The documentation is very consistent about calling it a NUL. I don't think "NUL=True" or "allow_NUL=true" is particularly attractive; we never (almost never?) use capital letters in parameter names. So any other name is going to be a compromise. "allow_null" and "allow_nul" are misspellings, and don't convey the idea any better; they can confuse the reader with the related concept of NULL or None. At least "zeroes" has the benefit of being an actual word, representing a related concept.
Will you be done bikeshedding soon? |
|
Date |
User |
Action |
Args |
2015-05-04 19:44:31 | larry | set | recipients:
+ larry, zach.ware, serhiy.storchaka |
2015-05-04 19:44:31 | larry | set | messageid: <1430768671.63.0.312833189219.issue24000@psf.upfronthosting.co.za> |
2015-05-04 19:44:31 | larry | link | issue24000 messages |
2015-05-04 19:44:31 | larry | create | |
|